Absolute ethanol (≥99.8%), 2-propanol (99.99%), 1-butanol (99.99%), castor oil, and Tween-60^® (99.99%) were purchased from Sigma-Aldrich. Methotrexate (working standards) was provided by Lahore Chemical & Pharmaceutical Works (Private) Limited, Pakistan (the molecular structures of MTX and Tween 60 are shown in Supplementary Figures S1, S2 in Supplementary Material, respectively). Deionized and double-distilled water (conductivity ≤1 μs/cm and viscosity 0.01 poise) was used for dilution and other experimental purposes.

Three new μE formulations were prepared using Tween-60 (HLB = 14.9), ethanol, 2-propanol, and 1-butanol at a constant ratio (1:1). To obtain a translucent μE, a surfactant mixture (S_mix) was placed in an ultrasonicator, followed by the addition of an appropriate amount of castor oil and further dilution with water using the titration method. The pseudoternary phase diagram was mapped using Tween-60 as the surfactant, castor oil as oil, and water as the aqueous phase (Figure 1). Ethanol, 2-propanol, and 1-butanol were used as co-surfactants in the preparation of μE-A, μE-B, and μE-C, respectively. The final composition of the optimal μE-A consists of Tween-60 (40%), ethanol (40%), castor oil (9%), and water (11%); the optimal μE-B composition consists of Tween-60 (38%), 2-propanol (38%), castor oil (9%), and water (15%); and the optimal μE-C composition consists of Tween-60 (41%), 1-butanol (41%), castor oil (8%), and water (10%). The morphological transformation of all systems from an oil-rich (w/o) system to a water-rich (o/w) system was accessed at a constant oil-to-surfactant ratio by selecting a water dilution line AB highlighted in the pseudoternary phase diagram.

Under continuous stirring, the maximum amount of MTX was loaded into the optimal systems as follows: 3.5 wt.% at pH 6.5, 4.8 wt.% at pH 6.9, and 4.2 wt.% at pH 6.3 for μE-A, μE-B, and μE-C, respectively. To maintain the temperature at 25°C ± 0.5°C, a Lauda M20 thermostatic water bath was used. All three μE systems (MTX-free and MTX-loaded) were stable and remained clear for 6°months. A rotary evaporator was used to obtain the MTX NPs. All μE formulations were freeze-dried, placing the colloidal solution in a freeze dryer (model: Alpha 1-2 LD Plus; Christ) to remove all water and organic substances and obtain fine and dried MTX nanopowder. The loading efficiency percentage (LE%) and entrapment efficiency percentage (EE%) of MTX were computed using the following formula: L E % = I n i t i a l   w e i g h t   o f   M T X − w e i g h t   o f   M T X   i n   s u p e r n a t a n t w e i g h t   o f   f o r m u l a t i o n × 100 , (1) E E % = I n i t i a l   w e i g h t   o f   M T X − w e i g h t   o f   M TX   i n   r e s i d u a l   l i q u i d s I n i t i a l   w e i g h t   o f   M X D × 100 . (2)

The stability and homogeneity of all MTX-free and MTX-loaded μE formulations were checked by centrifugation of the samples for 15 min at 3,000 rpm using a Hermle Z-200A Centrifuge (Germany). The biological microscope (LABOMED FLR Lx 400, Jenoptik, Germany) had a magnification of 4×/10×/40×/100× and was used to manifest any type of structural transition occurring in the structure of MTX-free μE systems.

The phase changes from w/o to o/w via a bicontinuous phase within the MTX-free μE were assessed using electrical conductivity measurements. A conductometer (ADWA AD3000, Hungary) was used to measure the electric conductivity (σ), whereas viscosity (ƞ) was measured using a calibrated Brookfield viscometer (LVDV-2T, United States) at 25ºC ± 1°C with 150 rpm by flushing and washing the viscosity at each measurement.

The average droplet size and zeta potential of MTX-free and MTX-loaded μE systems were determined at room temperature using a Zetasizer (Malvern, Nano ZSP, United Kingdom). Morphological analyses were conducted to explore the internal structure of both MTX-free and MTX-loaded μEs. These investigations were performed using a JEOL 2100 high-resolution transmission electron microscope (HRTEM, Japan), equipped with a LaB6 electron gun operating at 200 KV. Samples of the MTX-free and MTX-loaded μE were carefully positioned onto copper grids and subsequently air-dried to prepare them for observation under the microscope.

FTIR spectra were recorded of pure MTX, MTX-free, and MTX-loaded μE systems at a resolution of 2 cm^-1 within the range of 500–4,000 cm^-1 using a Bruker FTIR (Alpha Series, Germany). Meanwhile, a spectrofluorophotometer (Shimadzu RF-6000, Japan) was used to measure steady-state fluorescence. A range of 300–600 nm was used to record the fluorescence spectra of MTX. These spectra were recorded in the aqueous phase, oil phase, and all S_mix (1:1) and optimal μE systems.

The dissolution rate of MTX-NPs was examined using USP apparatus I (Basket) at 50 rpm for 30 min in 900 mL of the dissolution medium. The temperature was maintained at 37°C ± 0.5°C. For the dissolution test, the PT-DT70 instrument was used, and the absorbance at 306 ± 2 nm was measured using a UV–visible spectrophotometer (Shimadzu UV-1800, Japan). After a set period of 45 min, the samples were removed, and the extracted liquid was replaced with the dissolution medium. The sample was filtered using a 0.45-µm filter paper. Each test was performed in triplicate.

A comparative dissolution profile of MTX-NPs and methotrexate–commercial formulation (MTX-CF) was also studied at buffer pH 1.2, buffer pH 4.5, and buffer pH 6.8. Twelve units (equivalent to the same active substance weight) of both types of products were taken. The dissolution time was 60 min, and the same conditions were applied as mentioned above for the study. The sample was drawn at 5, 10, 15, 20, 30, 45, and 60 min. The sample was drawn and filtered using a 0.45-µm filter paper. The same volume was added to the dissolution as drawn. The UV–visible spectrophotometer (Shimadzu UV-1800) was used to measure the sample and standard solution at 306 ± 2 nm.

MTX-NP-μE-A, MTX-NP-μE-B, and MTX-NP-μE-C were stored at 15°C and elevated temperatures (40°C) for more than 6 months. No change in physical form was observed for all three formulations. The stability data show that all three formulations are stable and can be used commercially.

A ternary phase diagram is used to help study the phase behavior and determine the ideal circumstances for the generation of μEs by checking the compatibility of oil, water, and the surfactant (Rahman et al., 2017). It also helps determine the correlation between the phase behavior of the excipients of μEs and drug molecules (Mitchell and Ninham, 1981). The water dilution method was used to investigate the behavior of each phase; it is a quick, precise, and economical procedure (Pal et al., 2017). Optimal μE-A comprises Tween-60 (40%), ethanol (40%), castor oil (9%), and water (11%); optimal μE-B comprises Tween-60 (38%), 2-propanol (38%), castor oil (9%), and water (15%); and optimal μE-C comprises Tween-60 (41%), 1-butanol (41%), castor oil (8%), and water (10%), respectively. The shaded area in Figure 1 shows the μE region, and the dilution line is indicated by red lines in that μE region. The highlighted mark on the dilution line signifies the optimal μE that is used for further investigation and characterization (Saleem et al., 2018). Water-rich quantity shows that the optimal μE is w/o μE. Figure 1 shows the differences in the μE region, which are attributed to the varying co-surfactants used in the three systems (Siddique et al., 2021a).

The compatibility of oil with the co-surfactant and surfactant chain length plays a crucial role in determining the formation of µE structures (Lawrence and Rees, 2012). The different physiochemical properties of the μE systems are presented in Table 1. However, the microstructural transitions in the structure of the μE system cannot be evaluated using the ternary phase diagram. Therefore, the transitions in the one-phase region of the µE system are explored by conductivity (σ), viscosity (η), and optical microscopic analysis as a weight fraction of the aqueous component (Φ_w).

Electrical conductivity is a useful technique for evaluating the structural transition and forecasting a conductive network channel (bicontinuous μE) in µEs. The conductance is measured along the dilution line AB by constantly adding water to the oil, surfactant, and co-surfactant mixture (Acosta et al., 1996; Yadav et al., 2018). As the water was added, a change in the electrical conductance of the mixture occurred, as shown in Figure 2, which displays the plot of σ and its first derivative (dσ/dΦ) versus Φ_w for each µE system. An abrupt change occurred when the phase transition occurred. The phase transition from a w/o to o/w µE occurred, although the bicontinuous phase is determined by the conductivity (σ) of the water component (weight fraction) Φ_w (Kahlweit et al., 1993; Olivieri et al., 2003).

For µE-A, Figure 2 shows that the Φ_w value is below ∼8%, and the bicontinuous region of µE-A begins at ∼ 9% Φ_w, called the percolation threshold (Φ_p), below the slight increase observed in the Φ_w value (w/o µE). At a value of 16% Φ_w, sudden changes occurred, which indicated that σ decreased due to the increase in water content. The increase in water content leads to the development of o/w, which leads to the phase transition of Φ_b. With the increase in the value of Φ_w, the change in the first derivative (dσ/dφ) also further helps determine the phase transition in the µE domain (Formariz et al., 2008; Pal et al., 2017).

For µE-B, Φ_w slightly increases until the critical Φ_w value is 4.5%. The bicontinuous region starts at 5% of the percolation threshold (Φ_p). At this value, Φ_p w/o µE exists and is higher than the Φ_p value (Φ_w > 4.4%). σ increases until Φ_b (Φ_w ∼8.6%) is formed. After Φ_b, the sudden decrease in the σ value corresponds to an increase in the water ratio, thereby forming the o/w µE (Gaudana et al., 2010). For µE-C, the Φp value starts at 6.0% Φ_w, which is slightly higher than the Φ_w value (w/o µE). The changes occur at the 14.5% Φ_w value, which indicates that σ decreases as the water content increases. The increase in water content causes the development of the o/w µE, which leads to the phase transition of φb

The biological microscope was used to examine the microstructural transitions in µE systems via the bicontinuous phase, which determines the process of microstructure modification of µE (Nazar et al., 2020). Figures 4A–C show three anticipated phase transitions of µE-A; w/o µE, bicontinuous networks, o/w µE, and the proposed microstructure changes are shown in Figure 4, which shows the microstructure transformation of µEs with increasing concentration of the aqueous phase (Paria and Khilar, 2004).

Figure 4A shows a w/o µE with dispersed water droplets in the oil phase, while the o/w µE revealed that the oil droplets were present in the continuous aqueous phase (Figure 4C), and the bicontinuous µE showed a network of spherical droplets creating bicontinuous channels (Figure 4B). These results were also consistent with earlier studies (Nazar et al., 2018). The hydrophilic–lipophilic balance (HLB) between the surfactant and co-surfactant had an impact on the microstructures of the µE systems. o/w systems were produced by a lipophilic-leaning HLB, whereas w/o systems were produced by a hydrophilic-leaning HLB. When neither oil nor water droplets predominated, a bicontinuous µE appeared, suggesting percolation behavior (Khan et al., 2016; Rahman et al., 2016). The microstructure transformation of µE-B and µE-C with increasing concentrations of the aqueous phase obtained from an optical microscope is shown in Supplementary Material (Supplementary Figure S3).

The enhanced solubility of MTX in each optimized μE formulation is achieved, i.e., 3.5 wt% at pH 6.5, 4.8 wt% at pH 6.9, and 4.2 wt% at pH 6.3 in μE-A, μE-B, and μE-C, respectively. Furthermore, high EE% and excessive LE% are obtained for each formulation. The quantitative EE% and LE% are 94.22% ± 0.48% and 22.50% ± 0.48 for μE-A, 86.78% ± 0.92% and 17.75% ± 0.48 for μE-B, and 82.45% ± 1.15% and 15.95% ± 0.48 for μE-C, respectively.

The evaluation of physical stability is a key factor in the preparation of µE systems. Particles having a smaller size exhibit a higher surface area and can easily permeate with fast release (Siddique et al., 2024). The particle size distribution of MTX-free µE-A and MTX-loaded µE-A was computed, as shown in Figure 5A. The average size of MTX-free µE-A was ∼42 nm, with a polydispersity index (PDI) of 0.112. Likewise, the average size of MTX-loaded µE-A was ∼62.5 nm, and the PDI was 0.194. The DLS results showed an increase in size upon the loading of MTX in the optimal µE, which confirmed the encapsulation of MTX (Zafar et al., 2024). MTX-loaded µE-A exhibited an increase in size due to the loading of MTX (Rahman et al., 2017; Siddique et al., 2021a). The small size of droplets provides higher mobility, enhanced surface area for encapsulation, and enhanced dissolution and maximum release of the drug. Moreover, the larger size of the µE systems provides less mobility, which leads to slow drug release (Nazar et al., 2018; Siddique et al., 2021b). The particle size distribution of MTX-free µE-B (∼55 nm), MTX-loaded µE-B (∼70.5 nm), MTX-free µE-C (∼57.5 nm), and MTX-loaded µE-C (∼72 nm) is given in Supplementary Material (Supplementary Figure S4A, B).

Zeta (ζ) potential is the most significant parameter to evaluate the stability of colloidal systems. A higher value of ζ-potential indicates higher stability without any aggregation of droplets, while a low zeta potential shows less stability, leading to aggregation or coagulation. A highly stable colloidal system shows a ζ-potential value of >30 mV or < -30 mV due to the steric and electrostatic repulsion between particles (Nazar et al., 2009; Nazar et al., 2018). These ζ-potential calculations show the stability of MTX-loaded µE-A (−36.95 mV), MTX-loaded µE-B (−28.6 mV), and MTX-loaded µE-C (−36.8 mV) in Supplementary Material (Supplementary Figure S5). The substantial negative ζ-potential values of nanodroplets in µE systems are very suitable for the development of a stable drug delivery system. Additionally, the higher negative ζ-potential value demonstrated the improved stability and longer shelf life of the µE system (Bhagyaraj and Krupa, 2020; Saleem et al., 2020; Freidus et al., 2021).

FTIR is a powerful analytical technique that provides information about the molecular structure and functional groups present in a sample. FTIR analysis was used to investigate possible interactions between MTX and the different components of the µE system (Baptista and Tran, 1997; Dinache et al., 2020). The investigation concentrated on finding characteristic infrared peaks and patterns, which can show any alterations or shifts indicating chemical interactions in any µE system before and after loading MTX. In order to guarantee the durability and efficacy of the µE system as a drug delivery mechanism, this procedure is essential (Nazar et al., 2009).

The chemical stability of MTX in µE systems and the interactions between MTX and the other components of µEs were evaluated. As shown in Figure 5B, the FTIR of MTX showed the following characteristic peaks: a major peak at 3,361 cm^-1 was observed due to the carboxylic group [O-H stretching], whereas the peak at 2,949 cm^-1 is assigned to the CH₃ group [C-H stretching]. The peaks at 1,639 cm^-1 were due to the stretching of the carbonyl group [C=O stretching] and aromatic rings [C=C stretching]. These spectroscopic studies of µE-A, µE-B, and µE-C, together with MTX-free and MTX-loaded µEs, are shown in Figure 5C and Supplementary Figure S4C, D in Supplementary Material, respectively. The MTX-loaded µE-A, µE-B, and µE-C showed that MTX was completely dissolved in optimal µE systems without any aggregation and absence of any additional peak. Hence, there were no observable interactions present between MTX and µE components, which confirms the chemical stability of MTX in the microstructure of the µE systems (Rajinikanth et al., 2007).

The TEM micrographs shown in Figure 6 offer a detailed view of the high-resolution morphology of both the MTX-free μE and MTX-loaded μE. These images confirm a monomodal size distribution consistent with the estimations from DLS studies. Importantly, they reveal a notable increase in the size of the MTX-loaded μE compared to the MTX-free μE, indicating successful MTX loading. The micrographs of the MTX-free μE depict spherical surfaces with fine distribution, suggesting minimal alteration in morphology across all nano-colloidal dispersions. However, the increase in size observed implies the effective encapsulation of MTX, facilitating enhanced bioavailability and solubility. Specifically, μEs are utilized to encapsulate drugs with poor solubility and limited absorption capacity. The formation of these aggregates relies on self-assembly structures or patterns that navigate various obstacles to reach specific destinations, enabling controlled drug release at targeted sites (Alswieleh et al., 2020; Hanafy et al., 2023).